“Uncanny Valley 2” is a study that examines adults’ beliefs and feelings about a collection of real-world robots based on a viewing of an 8-second video of that robot.

There are 473 participants.

11 did not enter birthdates. 0 were too young.

Gender breakdown before exclusion
Another gender identity I prefer not to answer this question Man Woman
3 5 264 190
Age
Min. 1st Qu. Median Mean 3rd Qu. Max.
20 29 34 36.35 42 75

There are 462 participants after removing anyone under the age of 18.

36 did not complete the study in a sufficient amount of time.

Duration before exclusion
Min. 1st Qu. Median Mean 3rd Qu. Max.
0.55 3.017 4.008 5.321 5.775 50.47
Duration after exclusion
Min. 1st Qu. Median Mean 3rd Qu. Max.
2.017 3.171 4.05 4.764 5.596 14.43

There are 426 particpants after removing anyone with a study duration outside of the range of 2 - 15 minutes.

Gender breakdown
Another gender identity I prefer not to answer this question Man Woman
3 5 240 178
Education breakdown (continued below)
College graduate Have not completed high school degree High school graduate
242 4 127
Postgraduate education
53

Descriptives

Analysis of Questions

Confirmatory Factor Analysis

## lavaan (0.5-23.1097) converged normally after  52 iterations
## 
##   Number of observations                          2079
## 
##   Estimator                                         ML
##   Minimum Function Test Statistic               60.350
##   Degrees of freedom                                11
##   P-value (Chi-square)                           0.000
## 
## Model test baseline model:
## 
##   Minimum Function Test Statistic             8629.754
##   Degrees of freedom                                21
##   P-value                                        0.000
## 
## User model versus baseline model:
## 
##   Comparative Fit Index (CFI)                    0.994
##   Tucker-Lewis Index (TLI)                       0.989
## 
## Loglikelihood and Information Criteria:
## 
##   Loglikelihood user model (H0)             -10492.652
##   Loglikelihood unrestricted model (H1)     -10462.477
## 
##   Number of free parameters                         17
##   Akaike (AIC)                               21019.304
##   Bayesian (BIC)                             21115.178
##   Sample-size adjusted Bayesian (BIC)        21061.168
## 
## Root Mean Square Error of Approximation:
## 
##   RMSEA                                          0.046
##   90 Percent Confidence Interval          0.035  0.058
##   P-value RMSEA <= 0.05                          0.672
## 
## Standardized Root Mean Square Residual:
## 
##   SRMR                                           0.021
## 
## Parameter Estimates:
## 
##   Information                                 Expected
##   Standard Errors                             Standard
## 
## Latent Variables:
##                    Estimate  Std.Err  z-value  P(>|z|)
##   agency =~                                           
##     choose            1.000                           
##     think             1.483    0.113   13.076    0.000
##   exp =~                                              
##     scared            1.000                           
##     pain              1.034    0.020   53.027    0.000
##     hungry            0.961    0.019   51.160    0.000
##   uv =~                                               
##     creepy            1.000                           
##     weird             1.345    0.102   13.244    0.000
## 
## Covariances:
##                    Estimate  Std.Err  z-value  P(>|z|)
##   agency ~~                                           
##     exp               0.084    0.008   10.044    0.000
##     uv                0.070    0.012    5.643    0.000
##   exp ~~                                              
##     uv                0.053    0.008    6.330    0.000
## 
## Variances:
##                    Estimate  Std.Err  z-value  P(>|z|)
##    .choose            0.657    0.029   22.584    0.000
##    .think            -0.091    0.046   -1.983    0.047
##    .scared            0.052    0.002   23.996    0.000
##    .pain              0.030    0.002   16.892    0.000
##    .hungry            0.036    0.002   20.821    0.000
##    .creepy            0.419    0.053    7.973    0.000
##    .weird            -0.195    0.092   -2.106    0.035
##     agency            0.302    0.030   10.065    0.000
##     exp               0.137    0.006   23.539    0.000
##     uv                0.691    0.060   11.495    0.000
## $lambda
##        agency   exp    uv
## choose  0.561 0.000 0.000
## think   1.076 0.000 0.000
## scared  0.000 0.851 0.000
## pain    0.000 0.911 0.000
## hungry  0.000 0.882 0.000
## creepy  0.000 0.000 0.789
## weird   0.000 0.000 1.088
## 
## $theta
##        choose think  scared pain   hungry creepy weird 
## choose  0.685                                          
## think   0.000 -0.158                                   
## scared  0.000  0.000  0.275                            
## pain    0.000  0.000  0.000  0.171                     
## hungry  0.000  0.000  0.000  0.000  0.222              
## creepy  0.000  0.000  0.000  0.000  0.000  0.378       
## weird   0.000  0.000  0.000  0.000  0.000  0.000 -0.184
## 
## $psi
##        agency exp   uv   
## agency 1.000             
## exp    0.415  1.000      
## uv     0.153  0.172 1.000

Partial Correlations

A Priori Variables

## 
##  Pearson's product-moment correlation
## 
## data:  RBI$exp.c and RBI$agency.c
## t = 16, df = 2100, p-value <0.0000000000000002
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
##  0.2931 0.3696
## sample estimates:
##    cor 
## 0.3319
## 
## Call:
## lm(formula = uv.c ~ exp.c + agency.c + Gender + Age, data = RBI)
## 
## Residuals:
##    Min     1Q Median     3Q    Max 
## -1.846 -0.746 -0.278  0.529  2.404 
## 
## Coefficients:
##             Estimate Std. Error t value      Pr(>|t|)    
## (Intercept) -0.62674    0.15237   -4.11 0.00004053128 ***
## exp.c        0.10228    0.02284    4.48 0.00000794687 ***
## agency.c     0.13517    0.02259    5.98 0.00000000256 ***
## Gender       0.24439    0.03974    6.15 0.00000000093 ***
## Age         -0.00559    0.00203   -2.75        0.0061 ** 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 0.97 on 2074 degrees of freedom
## Multiple R-squared:  0.0613, Adjusted R-squared:  0.0595 
## F-statistic: 33.8 on 4 and 2074 DF,  p-value: <0.0000000000000002
## 
## Call:
## lm(formula = uv.c ~ exp.c + agency.c + Gender + Age, data = RBI)
## 
## Standardized Coefficients::
## (Intercept)       exp.c    agency.c      Gender         Age 
##     0.00000     0.10228     0.13517     0.13128    -0.05935

K-means clustering of a priori variables

## [1] 347.1
HAHE HALE LALE
104 659 1316
## 
## Call:
## lm(formula = uv ~ cluster.name, data = km)
## 
## Residuals:
##    Min     1Q Median     3Q    Max 
## -1.510 -0.774 -0.274  0.490  2.226 
## 
## Coefficients:
##                  Estimate Std. Error t value             Pr(>|t|)    
## (Intercept)        2.5096     0.0968   25.91 < 0.0000000000000002 ***
## cluster.nameHALE  -0.4990     0.1042   -4.79     0.00000179874413 ***
## cluster.nameLALE  -0.7357     0.1006   -7.31     0.00000000000037 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 0.988 on 2076 degrees of freedom
## Multiple R-squared:  0.0321, Adjusted R-squared:  0.0312 
## F-statistic: 34.5 on 2 and 2076 DF,  p-value: 0.00000000000000191
## 
## Call:
## lm(formula = uv ~ cluster.name, data = km)
## 
## Standardized Coefficients::
##      (Intercept) cluster.nameHALE cluster.nameLALE 
##           0.0000          -0.2315          -0.3535

Distribution of robots among k-means clusters

##          
##           1 2 3
##   atlas   1 0 0
##   spot    1 0 0
##   festo   0 1 0
##   kb      0 1 0
##   kf      0 1 0
##   nao     0 1 0
##   pepper  0 1 0
##   tapia   0 1 0
##   actroid 0 0 1
##   sofia   0 0 1
## 
## Call:
## lm(formula = uv ~ agency.c + exp.c + robot.group + Gender + Age, 
##     data = RBI)
## 
## Residuals:
##    Min     1Q Median     3Q    Max 
## -2.368 -0.641 -0.333  0.511  2.570 
## 
## Coefficients:
##                       Estimate Std. Error t value             Pr(>|t|)    
## (Intercept)            1.19708    0.14687    8.15  0.00000000000000062 ***
## agency.c               0.10475    0.02158    4.85  0.00000129410922935 ***
## exp.c                  0.08973    0.02127    4.22  0.00002553597944144 ***
## robot.grouphuman-like  0.74756    0.05953   12.56 < 0.0000000000000002 ***
## robot.grouprobotic    -0.18159    0.05083   -3.57              0.00036 ***
## Gender                 0.24621    0.03690    6.67  0.00000000003211117 ***
## Age                   -0.00586    0.00189   -3.10              0.00197 ** 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 0.9 on 2072 degrees of freedom
## Multiple R-squared:  0.197,  Adjusted R-squared:  0.195 
## F-statistic: 84.8 on 6 and 2072 DF,  p-value: <0.0000000000000002
## 
## Call:
## lm(formula = uv ~ agency.c + exp.c + robot.group + Gender + Age, 
##     data = RBI)
## 
## Standardized Coefficients::
##           (Intercept)              agency.c                 exp.c 
##               0.00000               0.10440               0.08943 
## robot.grouphuman-like    robot.grouprobotic                Gender 
##               0.30981              -0.09000               0.13181 
##                   Age 
##              -0.06202

Distribution of participants among k-means clusters

Number of participants that fall into 1, 2, or 3 clusters
1 2 3
254 158 9
Number of participants that fall into only one cluster
0 HAHE HALE LALE
167 11 58 185
Number of participants that fall into only two clusters
0 HALE & HAHE HALE & LALE LALE & HAHE
263 5 147 6

Data-driven aggregates

Principal Components Analysis

##            PC1      PC2     PC3     PC4     PC5     PC6      PC7
## choose 0.07515 0.353921 0.18439 0.21204 0.08988 0.01309 0.005868
## feel   0.16166 0.067631 0.04327 0.13742 0.38383 0.01152 0.052400
## hungry 0.16202 0.105847 0.16455 0.08313 0.07457 0.26062 0.348759
## moral  0.15484 0.008355 0.34667 0.15482 0.11675 0.16914 0.012940
## pain   0.16465 0.095220 0.13301 0.09173 0.13239 0.05776 0.428914
## scared 0.16425 0.097492 0.01472 0.01464 0.09630 0.45947 0.136773
## think  0.11742 0.271536 0.11339 0.30623 0.10628 0.02840 0.014346
## Importance of components:
##                          PC1   PC2    PC3    PC4    PC5    PC6    PC7
## Standard deviation     2.061 1.127 0.6504 0.5922 0.5313 0.4860 0.4358
## Proportion of Variance 0.607 0.182 0.0604 0.0501 0.0403 0.0337 0.0271
## Cumulative Proportion  0.607 0.788 0.8487 0.8988 0.9391 0.9729 1.0000

## 
##  Pearson's product-moment correlation
## 
## data:  pca$PC1 and pca$PC2
## t = 0.00000000000031, df = 2100, p-value = 1
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
##  -0.04299  0.04299
## sample estimates:
##                 cor 
## 0.00000000000000677
## 
## Call:
## lm(formula = uv ~ PC1 + PC2 + robot.group + Gender + Age, data = pca)
## 
## Residuals:
##    Min     1Q Median     3Q    Max 
## -2.360 -0.645 -0.319  0.508  2.565 
## 
## Coefficients:
##                       Estimate Std. Error t value             Pr(>|t|)    
## (Intercept)            1.20367    0.14705    8.19  0.00000000000000047 ***
## PC1                    0.06657    0.00983    6.78  0.00000000001612157 ***
## PC2                    0.05771    0.01803    3.20              0.00139 ** 
## robot.grouphuman-like  0.74069    0.05966   12.42 < 0.0000000000000002 ***
## robot.grouprobotic    -0.18386    0.05089   -3.61              0.00031 ***
## Gender                 0.24468    0.03693    6.62  0.00000000004418959 ***
## Age                   -0.00582    0.00190   -3.07              0.00220 ** 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 0.902 on 2072 degrees of freedom
## Multiple R-squared:  0.195,  Adjusted R-squared:  0.193 
## F-statistic: 83.6 on 6 and 2072 DF,  p-value: <0.0000000000000002
## 
## Call:
## lm(formula = uv ~ PC1 + PC2 + robot.group + Gender + Age, data = pca)
## 
## Standardized Coefficients::
##           (Intercept)                   PC1                   PC2 
##               0.00000               0.13673               0.06483 
## robot.grouphuman-like    robot.grouprobotic                Gender 
##               0.30697              -0.09112               0.13100 
##                   Age 
##              -0.06162

K-means clustering of a data-driven components

## [1] 1988
HAHE HALE LALE
114 658 1307

Distribution of robots among k-means clusters (PCA)

##          
##           1 2 3
##   actroid 1 0 0
##   sofia   1 0 0
##   festo   0 1 0
##   kb      0 1 0
##   kf      0 1 0
##   nao     0 1 0
##   pepper  0 1 0
##   tapia   0 1 0
##   atlas   0 0 1
##   spot    0 0 1
## 
## Call:
## lm(formula = uv ~ cluster, data = km)
## 
## Residuals:
##    Min     1Q Median     3Q    Max 
## -1.456 -0.771 -0.271  0.544  2.229 
## 
## Coefficients:
##             Estimate Std. Error t value             Pr(>|t|)    
## (Intercept)   2.0152     0.0385   52.32 < 0.0000000000000002 ***
## cluster2      0.4409     0.1002    4.40           0.00001143 ***
## cluster3     -0.2443     0.0472   -5.17           0.00000025 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 0.988 on 2076 degrees of freedom
## Multiple R-squared:  0.0312, Adjusted R-squared:  0.0303 
## F-statistic: 33.5 on 2 and 2076 DF,  p-value: 0.00000000000000489
## 
## Call:
## lm(formula = uv ~ cluster, data = km)
## 
## Standardized Coefficients::
## (Intercept)    cluster2    cluster3 
##      0.0000      0.1001     -0.1177

Distribution of participants among k-means clusters (PCA)

Number of participants that fall into 1, 2, or 3 clusters
1 2 3
255 156 10
Number of participants that fall into only one cluster
0 1 2 3
166 58 14 183
Number of participants that fall into only two clusters
0 1 & 2 1 & 3 2 & 3
265 6 147 3

Imputed Data

Analysis of Questions

Exploratory/Confirmatory Factor Analysis

## lavaan (0.5-23.1097) converged normally after  55 iterations
## 
##   Number of observations                          4260
## 
##   Estimator                                         ML
##   Minimum Function Test Statistic               90.078
##   Degrees of freedom                                11
##   P-value (Chi-square)                           0.000
## 
## Model test baseline model:
## 
##   Minimum Function Test Statistic            21151.329
##   Degrees of freedom                                21
##   P-value                                        0.000
## 
## User model versus baseline model:
## 
##   Comparative Fit Index (CFI)                    0.996
##   Tucker-Lewis Index (TLI)                       0.993
## 
## Loglikelihood and Information Criteria:
## 
##   Loglikelihood user model (H0)             -17960.535
##   Loglikelihood unrestricted model (H1)     -17915.496
## 
##   Number of free parameters                         17
##   Akaike (AIC)                               35955.069
##   Bayesian (BIC)                             36063.139
##   Sample-size adjusted Bayesian (BIC)        36009.120
## 
## Root Mean Square Error of Approximation:
## 
##   RMSEA                                          0.041
##   90 Percent Confidence Interval          0.033  0.049
##   P-value RMSEA <= 0.05                          0.966
## 
## Standardized Root Mean Square Residual:
## 
##   SRMR                                           0.020
## 
## Parameter Estimates:
## 
##   Information                                 Expected
##   Standard Errors                             Standard
## 
## Latent Variables:
##                    Estimate  Std.Err  z-value  P(>|z|)
##   agency =~                                           
##     choose            1.000                           
##     think             1.413    0.071   19.800    0.000
##   exp =~                                              
##     scared            1.000                           
##     pain              1.006    0.011   91.842    0.000
##     hungry            0.924    0.010   89.108    0.000
##   uv =~                                               
##     creepy            1.000                           
##     weird             1.183    0.042   28.221    0.000
## 
## Covariances:
##                    Estimate  Std.Err  z-value  P(>|z|)
##   agency ~~                                           
##     exp               0.080    0.005   14.994    0.000
##     uv                0.088    0.009    9.593    0.000
##   exp ~~                                              
##     uv                0.057    0.006   10.377    0.000
## 
## Variances:
##                    Estimate  Std.Err  z-value  P(>|z|)
##    .choose            0.610    0.019   32.456    0.000
##    .think            -0.084    0.027   -3.143    0.002
##    .scared            0.030    0.001   31.990    0.000
##    .pain              0.022    0.001   26.552    0.000
##    .hungry            0.023    0.001   30.048    0.000
##    .creepy            0.271    0.029    9.470    0.000
##    .weird            -0.088    0.039   -2.237    0.025
##     agency            0.296    0.020   15.017    0.000
##     exp               0.124    0.003   37.069    0.000
##     uv                0.814    0.036   22.518    0.000
## $lambda
##        agency   exp    uv
## choose  0.572 0.000 0.000
## think   1.080 0.000 0.000
## scared  0.000 0.897 0.000
## pain    0.000 0.922 0.000
## hungry  0.000 0.907 0.000
## creepy  0.000 0.000 0.866
## weird   0.000 0.000 1.041
## 
## $theta
##        choose think  scared pain   hungry creepy weird 
## choose  0.673                                          
## think   0.000 -0.166                                   
## scared  0.000  0.000  0.196                            
## pain    0.000  0.000  0.000  0.150                     
## hungry  0.000  0.000  0.000  0.000  0.178              
## creepy  0.000  0.000  0.000  0.000  0.000  0.250       
## weird   0.000  0.000  0.000  0.000  0.000  0.000 -0.084
## 
## $psi
##        agency exp   uv   
## agency 1.000             
## exp    0.417  1.000      
## uv     0.179  0.181 1.000

Partial Correlations

A Priori Variables

## 
##  Pearson's product-moment correlation
## 
## data:  RBI.imp$exp.c and RBI.imp$agency.c
## t = 36, df = 4300, p-value <0.0000000000000002
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
##  0.4617 0.5076
## sample estimates:
##   cor 
## 0.485
## 
## Call:
## lm(formula = uv.c ~ exp.c + agency.c + Gender + Age, data = RBI.imp)
## 
## Residuals:
##    Min     1Q Median     3Q    Max 
## -2.088 -0.744 -0.233  0.419  2.504 
## 
## Coefficients:
##                                            Estimate Std. Error t value
## (Intercept)                                -0.32943    0.18177   -1.81
## exp.c                                       0.08161    0.01705    4.79
## agency.c                                    0.16346    0.01699    9.62
## GenderI prefer not to answer this question  0.12543    0.22346    0.56
## GenderMan                                   0.45487    0.17780    2.56
## GenderWoman                                 0.66798    0.17819    3.75
## Age                                        -0.00567    0.00142   -3.99
##                                                        Pr(>|t|)    
## (Intercept)                                             0.07001 .  
## exp.c                                                 0.0000017 ***
## agency.c                                   < 0.0000000000000002 ***
## GenderI prefer not to answer this question              0.57463    
## GenderMan                                               0.01055 *  
## GenderWoman                                             0.00018 ***
## Age                                                   0.0000670 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 0.966 on 4253 degrees of freedom
## Multiple R-squared:  0.068,  Adjusted R-squared:  0.0667 
## F-statistic: 51.8 on 6 and 4253 DF,  p-value: <0.0000000000000002

K-means clustering of a priori variables

## [1] 377.6

Distribution of robots among k-means clusters

##          
##           1 2 3
##   festo   1 0 0
##   kb      1 0 0
##   kf      1 0 0
##   nao     1 0 0
##   pepper  1 0 0
##   tapia   1 0 0
##   actroid 0 1 0
##   sofia   0 1 0
##   atlas   0 0 1
##   spot    0 0 1
## 
## Call:
## lm(formula = uv ~ cluster, data = km)
## 
## Residuals:
##    Min     1Q Median     3Q    Max 
## -1.673 -0.762 -0.262  0.327  2.238 
## 
## Coefficients:
##             Estimate Std. Error t value            Pr(>|t|)    
## (Intercept)   1.7621     0.0190   92.86 <0.0000000000000002 ***
## cluster2      0.2771     0.0326    8.51 <0.0000000000000002 ***
## cluster3      0.9106     0.0791   11.51 <0.0000000000000002 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 0.987 on 4257 degrees of freedom
## Multiple R-squared:  0.0408, Adjusted R-squared:  0.0403 
## F-statistic: 90.5 on 2 and 4257 DF,  p-value: <0.0000000000000002
## 
## Call:
## lm(formula = uv ~ cluster, data = km)
## 
## Standardized Coefficients::
## (Intercept)    cluster2    cluster3 
##      0.0000      0.1290      0.1744
## 
## Call:
## lm(formula = uv ~ agency.c + exp.c + robot.group + Gender + Age, 
##     data = RBI)
## 
## Residuals:
##    Min     1Q Median     3Q    Max 
## -2.666 -0.648 -0.350  0.468  2.565 
## 
## Coefficients:
##                                            Estimate Std. Error t value
## (Intercept)                                 1.44516    0.17418    8.30
## agency.c                                    0.12902    0.01637    7.88
## exp.c                                       0.08757    0.01615    5.42
## robot.grouphuman-like                       0.81123    0.04426   18.33
## robot.grouprobotic                         -0.05352    0.03672   -1.46
## GenderI prefer not to answer this question  0.10925    0.21102    0.52
## GenderMan                                   0.45009    0.16790    2.68
## GenderWoman                                 0.66693    0.16827    3.96
## Age                                        -0.00605    0.00134   -4.50
##                                                        Pr(>|t|)    
## (Intercept)                                < 0.0000000000000002 ***
## agency.c                                     0.0000000000000041 ***
## exp.c                                        0.0000000622867409 ***
## robot.grouphuman-like                      < 0.0000000000000002 ***
## robot.grouprobotic                                       0.1450    
## GenderI prefer not to answer this question               0.6047    
## GenderMan                                                0.0074 ** 
## GenderWoman                                  0.0000750464852556 ***
## Age                                          0.0000068754133205 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 0.912 on 4251 degrees of freedom
## Multiple R-squared:  0.182,  Adjusted R-squared:  0.18 
## F-statistic:  118 on 8 and 4251 DF,  p-value: <0.0000000000000002
## 
## Call:
## lm(formula = uv ~ agency.c + exp.c + robot.group + Gender + Age, 
##     data = RBI)
## 
## Standardized Coefficients::
##                                (Intercept) 
##                                    0.00000 
##                                   agency.c 
##                                    0.12807 
##                                      exp.c 
##                                    0.08692 
##                      robot.grouphuman-like 
##                                    0.32213 
##                         robot.grouprobotic 
##                                   -0.02603 
## GenderI prefer not to answer this question 
##                                    0.01168 
##                                  GenderMan 
##                                    0.22160 
##                                GenderWoman 
##                                    0.32653 
##                                        Age 
##                                   -0.06338

Distribution of participants among k-means clusters

Number of participants that fall into 1, 2, or 3 clusters
1 2 3
229 185 12
Number of participants that fall into only one cluster
0 1 2 3
197 173 49 7
Number of participants that fall into only two clusters
0 1 & 2 1 & 3 2 & 3
241 173 1 11

Data-driven aggregates

Principal Components Analysis

##           PC1      PC2       PC3     PC4     PC5     PC6      PC7
## choose 0.0747 0.358496 0.2117172 0.18468 0.08039 0.01209 0.002792
## feel   0.1603 0.059840 0.0001788 0.15381 0.39083 0.01630 0.004330
## hungry 0.1622 0.105589 0.1611379 0.09788 0.07380 0.27867 0.362128
## moral  0.1551 0.003599 0.3181355 0.17725 0.18024 0.12411 0.001640
## pain   0.1648 0.098904 0.1217129 0.08539 0.08005 0.08309 0.489504
## scared 0.1659 0.098919 0.0324377 0.01673 0.09292 0.47091 0.127911
## think  0.1169 0.274654 0.1546800 0.28427 0.10178 0.01485 0.011694
## Importance of components:
##                          PC1   PC2    PC3    PC4    PC5    PC6    PC7
## Standard deviation     2.097 1.126 0.6341 0.5689 0.5215 0.4198 0.4004
## Proportion of Variance 0.628 0.181 0.0574 0.0462 0.0389 0.0252 0.0229
## Cumulative Proportion  0.628 0.809 0.8668 0.9131 0.9519 0.9771 1.0000

## 
## Call:
## lm(formula = uv ~ PC1 + PC2 + robot.group, data = pca)
## 
## Residuals:
##    Min     1Q Median     3Q    Max 
## -2.725 -0.623 -0.434  0.422  2.422 
## 
## Coefficients:
##                       Estimate Std. Error t value             Pr(>|t|)    
## (Intercept)            1.75241    0.03196   54.84 < 0.0000000000000002 ***
## PC1                    0.08502    0.00676   12.57 < 0.0000000000000002 ***
## PC2                    0.07760    0.01277    6.08         0.0000000013 ***
## robot.grouphuman-like  0.81115    0.04481   18.10 < 0.0000000000000002 ***
## robot.grouprobotic    -0.04475    0.03715   -1.20                 0.23    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 0.922 on 4255 degrees of freedom
## Multiple R-squared:  0.164,  Adjusted R-squared:  0.163 
## F-statistic:  208 on 4 and 4255 DF,  p-value: <0.0000000000000002
## 
## Call:
## lm(formula = uv ~ PC1 + PC2 + robot.group, data = pca)
## 
## Standardized Coefficients::
##           (Intercept)                   PC1                   PC2 
##               0.00000               0.17696               0.08676 
## robot.grouphuman-like    robot.grouprobotic 
##               0.32209              -0.02176
## 
##  Pearson's product-moment correlation
## 
## data:  pca$PC1 and pca$PC2
## t = -0.0000000000013, df = 4300, p-value = 1
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
##  -0.03003  0.03003
## sample estimates:
##                  cor 
## -0.00000000000001979
## 
## Call:
## lm(formula = uv ~ PC1 + PC2 + robot.group + Gender + Age, data = pca)
## 
## Residuals:
##    Min     1Q Median     3Q    Max 
## -2.650 -0.651 -0.343  0.476  2.567 
## 
## Coefficients:
##                                            Estimate Std. Error t value
## (Intercept)                                 1.43476    0.17423    8.23
## PC1                                         0.08064    0.00679   11.88
## PC2                                         0.07246    0.01267    5.72
## robot.grouphuman-like                       0.81046    0.04435   18.27
## robot.grouprobotic                         -0.04871    0.03677   -1.32
## GenderI prefer not to answer this question  0.11889    0.21107    0.56
## GenderMan                                   0.45854    0.16796    2.73
## GenderWoman                                 0.67248    0.16831    4.00
## Age                                        -0.00603    0.00135   -4.49
##                                                        Pr(>|t|)    
## (Intercept)                                 0.00000000000000024 ***
## PC1                                        < 0.0000000000000002 ***
## PC2                                         0.00000001159271421 ***
## robot.grouphuman-like                      < 0.0000000000000002 ***
## robot.grouprobotic                                       0.1854    
## GenderI prefer not to answer this question               0.5733    
## GenderMan                                                0.0064 ** 
## GenderWoman                                 0.00006567057731519 ***
## Age                                         0.00000744440065800 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 0.912 on 4251 degrees of freedom
## Multiple R-squared:  0.182,  Adjusted R-squared:  0.18 
## F-statistic:  118 on 8 and 4251 DF,  p-value: <0.0000000000000002
## 
## Call:
## lm(formula = uv ~ PC1 + PC2 + robot.group + Gender + Age, data = pca)
## 
## Standardized Coefficients::
##                                (Intercept) 
##                                    0.00000 
##                                        PC1 
##                                    0.16784 
##                                        PC2 
##                                    0.08101 
##                      robot.grouphuman-like 
##                                    0.32182 
##                         robot.grouprobotic 
##                                   -0.02369 
## GenderI prefer not to answer this question 
##                                    0.01271 
##                                  GenderMan 
##                                    0.22576 
##                                GenderWoman 
##                                    0.32925 
##                                        Age 
##                                   -0.06325

K-means clustering of a data-driven components

## [1] 3672

Distribution of robots among k-means clusters (PCA)

##          
##           1 2 3
##   actroid 1 0 0
##   sofia   1 0 0
##   festo   0 1 0
##   kb      0 1 0
##   kf      0 1 0
##   nao     0 1 0
##   pepper  0 1 0
##   tapia   0 1 0
##   atlas   0 0 1
##   spot    0 0 1
## 
## Call:
## lm(formula = uv ~ cluster, data = km)
## 
## Residuals:
##    Min     1Q Median     3Q    Max 
## -1.618 -0.762 -0.262  0.382  2.238 
## 
## Coefficients:
##             Estimate Std. Error t value             Pr(>|t|)    
## (Intercept)   2.6183     0.0724   36.17 < 0.0000000000000002 ***
## cluster2     -0.8561     0.0748  -11.44 < 0.0000000000000002 ***
## cluster3     -0.5805     0.0772   -7.52    0.000000000000065 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 0.987 on 4257 degrees of freedom
## Multiple R-squared:  0.0399, Adjusted R-squared:  0.0395 
## F-statistic: 88.6 on 2 and 4257 DF,  p-value: <0.0000000000000002
## 
## Call:
## lm(formula = uv ~ cluster, data = km)
## 
## Standardized Coefficients::
## (Intercept)    cluster2    cluster3 
##      0.0000     -0.4089     -0.2689

Distribution of participants among k-means clusters (PCA)

Number of participants that fall into 1, 2, or 3 clusters
1 2 3
231 181 14
Number of participants that fall into only one cluster
0 1 2 3
195 9 174 48
Number of participants that fall into only two clusters
0 1 & 2 1 & 3 2 & 3
245 3 8 170

Unfolding analysis

Latent Class Analysis